{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7BC6W4AOFOTO3DJ3DT6I2QHSKQ","short_pith_number":"pith:7BC6W4AO","schema_version":"1.0","canonical_sha256":"f845eb700e2ba6ed8d3b1cfc8d40f25429250cdc000955b0634d37dc20f5d213","source":{"kind":"arxiv","id":"2606.05177","version":1},"attestation_state":"computed","paper":{"title":"MCBench: A Multicontext Safety Assessment Benchmark for Omni Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.CL","authors_text":"Amar Kaur, Dinh Phung, Gholamreza Haffari, Junae Kim, Lizhen Qu, Manh Luong, Rollin Omari, Tamas Abraham, Trang Vu","submitted_at":"2026-04-17T12:31:17Z","abstract_excerpt":"Existing multimodal safety benchmarks focus solely on visual inputs and cannot assess Omni Large Language Models (LLMs) that process vision, audio, and text. We introduce MCBench, a benchmark with 1196 scenarios spanning four safety categories that require integrating multiple modalities for accurate safety assessment. Each unsafe scenario is paired with a minimally different safe counterpart to assess model sensitivity. Our evaluations of state-of-the-art models reveal significant challenges. Omni LLMs struggle with subtle or non-physical risks but perform better when salient visual or acoust"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.05177","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-17T12:31:17Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"f9252125f8a9517b91ff178c7ddf10451dc6a9fd0a6a86cd584ff82e319242ff","abstract_canon_sha256":"3b212c4a3a9143bba8b8758bfcf54daca585ba9123536b947f6f9a62b96f4bfb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T00:13:47.852382Z","signature_b64":"Esc8HMB8nhIaiyFvEw+VbbpDpzxkgf2p0fRvPkZ6cwK1wwRoabxlp6xkRn2N2oeuf48oO4MoBRWSGUwLeBgkDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f845eb700e2ba6ed8d3b1cfc8d40f25429250cdc000955b0634d37dc20f5d213","last_reissued_at":"2026-06-05T00:13:47.851767Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T00:13:47.851767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MCBench: A Multicontext Safety Assessment Benchmark for Omni Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.CL","authors_text":"Amar Kaur, Dinh Phung, Gholamreza Haffari, Junae Kim, Lizhen Qu, Manh Luong, Rollin Omari, Tamas Abraham, Trang Vu","submitted_at":"2026-04-17T12:31:17Z","abstract_excerpt":"Existing multimodal safety benchmarks focus solely on visual inputs and cannot assess Omni Large Language Models (LLMs) that process vision, audio, and text. We introduce MCBench, a benchmark with 1196 scenarios spanning four safety categories that require integrating multiple modalities for accurate safety assessment. Each unsafe scenario is paired with a minimally different safe counterpart to assess model sensitivity. Our evaluations of state-of-the-art models reveal significant challenges. Omni LLMs struggle with subtle or non-physical risks but perform better when salient visual or acoust"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05177","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.05177/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.05177","created_at":"2026-06-05T00:13:47.851862+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05177v1","created_at":"2026-06-05T00:13:47.851862+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05177","created_at":"2026-06-05T00:13:47.851862+00:00"},{"alias_kind":"pith_short_12","alias_value":"7BC6W4AOFOTO","created_at":"2026-06-05T00:13:47.851862+00:00"},{"alias_kind":"pith_short_16","alias_value":"7BC6W4AOFOTO3DJ3","created_at":"2026-06-05T00:13:47.851862+00:00"},{"alias_kind":"pith_short_8","alias_value":"7BC6W4AO","created_at":"2026-06-05T00:13:47.851862+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ","json":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ.json","graph_json":"https://pith.science/api/pith-number/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/graph.json","events_json":"https://pith.science/api/pith-number/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/events.json","paper":"https://pith.science/paper/7BC6W4AO"},"agent_actions":{"view_html":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ","download_json":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ.json","view_paper":"https://pith.science/paper/7BC6W4AO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05177&json=true","fetch_graph":"https://pith.science/api/pith-number/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/graph.json","fetch_events":"https://pith.science/api/pith-number/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/action/storage_attestation","attest_author":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/action/author_attestation","sign_citation":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/action/citation_signature","submit_replication":"https://pith.science/pith/7BC6W4AOFOTO3DJ3DT6I2QHSKQ/action/replication_record"}},"created_at":"2026-06-05T00:13:47.851862+00:00","updated_at":"2026-06-05T00:13:47.851862+00:00"}